The human polyomavirus, JCV, causes a demyelinating disease in immune compromised individuals, progressive multifocal leukoencephalopathy (PML). PML is a substantial neurological complication in AIDS patients, occuring in 8% of all cases. PML is the AIDS defining illness in almost 1 to 3% of AIDS cases. Although in the past PML was a rare disease, its incidence is high in immune compromised patients like AIDS and allograft recipients. JC Virus was orignially thought to be strictly neurotropic, infecting only macroglial cells derived from the human brain. However, our results have now shown that there is a firm link between susceptible lymphoid and glial cells in the pathogenesis of JCV infection leading to demyelination in the human brain. The data indicate that a similar viral genotype which is found in brain tissue of PML patients originates from an initial lymphoid cell infection, including the bone marrow. The molecular similarities for JCV gene expression between cells of the nervous and immune systems directly correlated with the expression of the NF-1 class X family of DNA binding proteins which function for viral transcription. We have cloned the gene for NF-1/X from human brain which is highly expressed. There is a close association between susceptibility of infection to JCV and expression of this DNA binding protein. We have made expression vectors for all four class members of the NF-1 family. We have made a unique human brain derived, multipotential progentior cell line that can differentiate to either neurons or glial cells. We have been able to over express NF-1/X in human neurons differentiated from human CNS progenitor cells that are not susceptible to infection since these cells do not express competent levels of NF-1/X. The NF-1/X over expressing neurons however are susceptible to JCV infection, confirming the necessary regulatory function of this transcription factor as we have shown in the past in hematopoietic cells. In addition, we have shown that the binding site for NF-1/X is directly adjacent to the c-jun binding site on the viral promoter. We have then shown that NF-1/X and c-jun proteins interact directly. Increased concentrations of c-jun can alter the ability of NF-1/X to bind to its cognate site particularly if c-jun is phosphorylated. We have also determined that JCV infection of human multiotential progenitor cells results in viral multiplication that is augmented 10 fold when differentiated to astrocytes but shut off when differentiated to neurons. New experiments have shown that infection can be reversed if the lineage is directed either back to progenitor cells or from glial cells to neurons. There is also an interplay among the class members of NF-1 since if increases of NF-1A in cultured cells diminishes JCV growth while its unique inhibition compared with NF-1X shows an increase in JCV growth. As a part of these studies, we have demonstrated that virus infection kills human glial cells as a lytic, necrotic cell death and not an apoptotic cell death as previously described by others. We have also been able to direct differentiation of the progenitor cells to an oligodendrocyte. However, these cells are not as susceptible to JCV productive infection as the astrocytes due in some measure to their lack of high levels of expression of DNA binding proteins that recognize the viral promoter. Also few of the oligodendroctes express the viral co-receptor, 5HT2A. In addition to these factors, we have identified another DNA binding protein, SpiB, that is upregulated in patients on therapies that are a risk for PML i.e. natalizumab. The SpiB protein functions in lymphoid cells as a regulator for B cell development and also binds sites in the JCV promoter. These sites are overexpressed in the nucleotide sequences that are found in the virus DNA in PML brain tissue while not found in the same number in non pathologic tissue like kidney. In addition we have mutated some of the sequences in the SpiB binding sites i.e. L5 and L4 that are found in the PML brain sequences. If the L5 site is mutated, the site near the functional TATA binding site near the viral or1 of replication, there is an augmentation of viral infection. If the L4 site is mutated, the site that is at the beginning of the second tandem repeat without a TATA site, then viral infection is substantially diminished. Further mutational analysis from PML brain tissues of the regulatory region has revealed that SpiB sites are duplicated but always conserved in multiple locations in the promoter/enhancer region of the genome. Also, although ther are similar binding sites in the genome just adjacent to the regulatory region, alterations in these sites do not effect viral gene expression. Importantly, mutation of the L3 site to a L5 site in the non-pathogenic sequence of the viral promoter derived from the urine allows expression of the viral T protein. This new observation helps explain the difference between viral multiplication of the non-pathogenic compared with the pathogenic variants. A single nucleotide change does effect viral gene expression. SpiB protein functions for B cell development and is expressed in glial cells in the human brain that is a novel observation from these studies. These observations provide insight into the pathogenesis of JCV infection using both immune and nervous system cells for viral multiplication. Further it is clear that SpiB is synthesized in CD19/CD20 cells in the peripheral circulation that correlates with JCV susceptible cell types. SpiB is a candidate molecular biomarker for risk of the development of PML.